Recommendations
0 to acquire · 1 to dispose
| Ticker | Name | Close | Action | Action Price Where to act — a buy-limit on a pullback or a buy-stop above resistance, not the current price. Sanity-check vs Close: on a held name, an action price stranded above the close means the queued level was set when the stock was higher — stale, awaiting reprice. | Stop The risk-defined exit, set below the entry for a long (~1.2–1.8×ATR). A held long whose Stop sits at/above the Close is already below its stop — a stale box awaiting reprice. | 1wk Model estimate of the likely price at this horizon — a forward projection, not a guarantee, and separate from the trade's Action Price or Stop. Where the ML estimator is enabled these are per-ticker calibrated forecasts (~80% coverage band) that are refreshed periodically and can drift between refreshes; otherwise they're interpolated from the signal's swing target. Confirm against the settled close. | 4wk Model estimate of the likely price at this horizon — a forward projection, not a guarantee, and separate from the trade's Action Price or Stop. Where the ML estimator is enabled these are per-ticker calibrated forecasts (~80% coverage band) that are refreshed periodically and can drift between refreshes; otherwise they're interpolated from the signal's swing target. Confirm against the settled close. | 12wk Model estimate of the likely price at this horizon — a forward projection, not a guarantee, and separate from the trade's Action Price or Stop. Where the ML estimator is enabled these are per-ticker calibrated forecasts (~80% coverage band) that are refreshed periodically and can drift between refreshes; otherwise they're interpolated from the signal's swing target. Confirm against the settled close. | Rationale |
|---|---|---|---|---|---|---|---|---|---|
| S63 | ST Engineering | 10.27 | SELL 50% | 10.27 | 10.01 | 10.26 (-0.1%) | 10.26 (-0.1%) | 10.17 (-1.0%) | We are taking half of ST Engineering off the table. The business is genuinely good - first-half profit rose 27% and the order book is at a record S7bn - but the share price has stalled well past what the fundamentals support, and every forward estimate we have is now flat-to-lower over one, four and twelve weeks. This is a de-risk of an expensive good business, not a verdict on the company, which is why we are selling half and keeping half. OPERATIONAL: THIRD NIGHT, CYCLE 2, LAST PERMITTED REPUBLICATION. Ordered 09-08 for the 09-09 open (not executed), re-ordered 09-09 as an unconditional SG-1 market order for the 09-10 open (not executed); input/sgportfolio.json still reads,800. On 09-11 this crosses SG-8's >=3-cycle bar and MUST convert from a decision card into a single blocking 'holdings not synced' flag - stated in advance. NON-EXECUTION HAS COST REAL MONEY: action price 10.37 -> 10.27 = Sof slippage on (Svs the original 09-08 card at 10.34) - exactly the bleed L3/SG-1 exist to stop. Exit case intact and slightly stronger: Est_12W 10.17 = -0.97% (was -0.96%); the 09-07 re-arm 'settled close <=10.39 with Est_12W negative' is satisfied at 10.27 with 12c of room. Engine silent (NO TRADE cv0, 'Consensus conflict + low conviction') per SG-2 - a CLAUDE.md SS9 / L2 exit, never label-driven. Below EMA20 10.5072, above EMA200 10.0262: a stall inside an intact long-term trend, hence 50% not 100%. Leg 1 does not print because the residual stop was deliberately re-based 10.51 -> 10.01 on 09-08 ON THE CONDITION that we act immediately; two sessions later price has fallen TOWARD that line, so acting is what keeps the re-basing legitimate - a third non-execution converts it into a stop simply walked away from. Portion 50% = 30% base amplified one tier under SG-3 on MoS_FV -9.6455; Quality_Pass TRUE so never 100%. Market order: 10.27 is a reference only, no limit to reprice; flag if the open gaps >~1.5%. Entry range 10.23-10.31 (+/-0.25x ATR 0.15378). Proceeds ~Sof a Sposition. Estimates (live Est_*, ~80% range): 1wk 10.26 (-0.10%) [9.82-10.75], 4wk 10.26 (-0.10%) [9.60-11.12], 12wk 10.17 (-0.97%) [8.11-12.96] - band skewed -21%/+26%. Est_12W exceeds DCF_per_share 6.831 - ANNOTATED, not modified, not used to amplify. Residual stop HELD 10.01 (2.0x ATR chandelier 9.962, EMA200 10.0262 both at/below the carried line - not lowered). Scorecard NoTrade|sell -1 (n=13, hit.455) noted; does not gate a SS9 exit. L8: executes at the 2026-09-11 open; horizons and scorecard grade from the EXECUTION date/price. · news |
| V03 | Venture Corporation | 16.74 | WATCH | 16.72 | 16.38 | 16.75 (+0.2%) | 16.75 (+0.2%) | 16.42 (-1.8%) | Venture is the best-looking buy on the board tonight and we are still not buying it. It is a proper pullback setup rather than a chase, but our own track record on this exact type of trade in this book is poor - one in three works - and the model's twelve-week view points 1.9% lower, so the setup is not strong enough to overcome its own history. We would need a clearly stronger signal before committing cash. OPERATIONAL: adj 55 = raw 60 + val_adj -2 (MoS -0.3687 x 5 = -1.84) + bias -3 (PullbackBuy_VWAP|buy n=6 hit_1w 0.333, ACTIONABLE) - TEN POINTS BELOW the 65 floor; the refusal needs no second argument. Second argument exists anyway: Est_12W 16.42 = -1.91%, DEEPER negative than the -1.54% that refused Z74 on 09-09 and the -0.67% that refused BN4 on 09-08. Signal genuine and non-extended (eD -0.07), Buy_Rank #1 - but a cv60 PullbackBuy in this book's worst actionable buy cell is exactly the trade the bias exists to stop. THIRD refusal, structural: V03 is one of two UNRECONCILED names (ADR-0019) - sg_state.json carries from the 08-31 fill, input/sgportfolio.json reads; we cannot size an add to a position whose existence our own files dispute. RE-ARM, numeric and fixed: a settled-close BUY reprint at conviction >= 70 (so raw 70 - 2 - 3 = 65 clears the floor WITH the live bias applied) AND Est_12W at or above +1.0% vs the entry arms the acquire at the next open, at the then-live Entry_Price, no carried limit. Note the -3 bias makes this book need cv70 on a PullbackBuy where a neutral cell needs 65. |
| U11 | United Overseas Bank | 40.99 | HOLD | 40.99 | 41.27⚠ | 40.94 (-0.1%) | 40.77 (-0.5%) | 40.29 (-1.7%) | UOB has slipped slightly below the level where we said we would consider trimming, and the twelve-week view has turned mildly negative - but it has not reached the specific price we published as our trigger, so we are holding. This is our largest position and the bank itself remains sound and reasonably valued; we will act on our stated line, not one session early. OPERATIONAL: L2's BOTH LEGS met for a second session - close 40.99 vs stop 41.27 = -0.68% = 0.55x ATR (0.5061), materially deeper than last night's 0.16x, and Est_12W 40.29 = -1.71% negative. REFUSED because the book's own published numeric re-arm did not fire: 09-09 set 'a settled close at or below 40.85' and 40.99 is 14c above it. Cutting S - the largest position on the book - one session after publishing a line price did not reach would destroy the only discipline currently working here: the S63 10.39 trigger fired and was acted on; the Z74 re-arm did not fire and the name collapsed. RE-ARM UNCHANGED: settled close at or below 40.85 with Est_12W still negative arms the L2 exit at the next open. Stop HELD 41.27, NOT lowered. Context: far above EMA200 37.97, just below EMA20 41.35; engine prints HOLD cv41 ('signal neutral/weak-bullish; no action'). HORIZON: the 09-11 US CPI bank leg sits directly on this name. · news |
| HMN | CAPITALAND and ASCOTT STAPLE U | 0.85 | HOLD | 0.85 | 0.88⚠ | 0.85 (+0.0%) | 0.85 (+0.0%) | 0.87 (+2.4%) | Ascott Trust is flashing a sell signal, but the model still expects it about 3% higher in twelve weeks and the position is tiny - roughly a thousand dollars - so selling it would be noise, not risk management. Interest rates turning is the sector's biggest tailwind in three years, and the trust was just added to the STI reserve list on a solid operating update. We hold. OPERATIONAL: the ONLY name in the S-REIT sleeve clearing the 65 floor tonight - adj 66 = raw 69 + bias -3 (RallySell_BearTrend|sell n=9 hit.333, ACTIONABLE, DEEPENED from -1 at the 07-18 review). Refused on SG-2 leg 2: Est_12W +2.96% positive. Position S; the 80.07% portion is S - economically indistinguishable from noise. 12th consecutive sleeve refusal, and tonight it got EASIER not harder: last night four names cleared the floor, tonight one. Sleeve conviction faded hard in one session (J69U, BUOU and OUT of the sell pool, N2IU - direct tape evidence for SG-2/SG-5 that SG sell labels reverse without price confirming them. RSI across the sleeve 28.7-43.4; selling here sells the oversold low into a model projecting a bounce. HORIZON corroborates trail-not-exit: 09-11 CPI (S-REIT leg), 09-16 FOMC, 09-18 BoJ all prescribe TIGHTEN-TRAIL. Stop HELD 0.88, not lowered; breach -3.98%. · news |
| O39 | OCBC Bank | 31.39 | HOLD | 31.39 | 31.30 | 31.39 (+0.0%) | 31.45 (+0.2%) | 32.81 (+4.5%) | OCBC is our best holding, up 21% on cost, and it has again hit the profit-taking threshold - but nothing about it is deteriorating, so we let it run rather than clip it. The twelve-week view actually improved this session and conviction is steady. OPERATIONAL: ADR-0012 re-check - ARMED (Gain_ATR 10.89, T12_Progress 95.67, +21.06%; the ONLY holding with a real Avg_Cost 25.93 per SG-7) with ZERO OF FOUR fire ticks: (a) PROFIT_ARMED prints alone, no second Exit_Warning code; (b) conviction flat (dipped to 46 intraday, recovered), no >=20pt collapse; (c) Est_12W 32.81 > 1.02 x 31.39 = 32.02, forward view ALIVE and IMPROVED from 32.45 last night; (d) HOLD, not NO TRADE. Armed-without-a-tick means trail-raise, not trim - and the raise is DECLINED because both candidate stops sit BELOW the carried line: 2.0x ATR chandelier 31.39 - 1.003 = 30.39, model stop 30.39, vs carried 31.30. Stop HELD 31.30, NOT lowered - still the tightest line on the book at 0.29% under the close. No breach. |
| M44U | Mapletree Logistics Trust | 1.13 | WATCH | 1.13 | 1.09 | 1.13 (+0.0%) | 1.15 (+1.8%) | 1.17 (+3.5%) | Mapletree Logistics is the one name we do not own that is steadily improving - it is deeply oversold, turning up, and carries the best twelve-week outlook of any non-holding on the board at +3.5%. It is not yet a buy signal, so we are watching rather than acting, but this is the first name in weeks that could become a clean entry. OPERATIONAL: conviction climbing three consecutive prints ->, ReversalUp_Oversold, RSI 34.5, Est_12W 1.17 = +3.54% - the strongest positive forward view on any non-held SG name tonight. NOT ACTIONABLE: WATCH is not in the buy pool, adj 58 is sub-floor (< 65), and valuation is NA (val_adj 0, quality gate off). Logged because a BUY reprint at cv >= 65 would be this book's first genuinely clean acquire in weeks. No position; entry 1.13, stop 1.09 are the model's levels for reference only. |
Outcome & track record
Accuracy at the time of this record.
130 Graded decisions Past recommendations old enough to score against what actually happened (target hit / stopped out / closed). This is the evidence base — the bigger the number, the more history behind the calls. Click to inspect every one. See the decisions →
0.162 1-week calibration How much of the eventual 4-week move typically shows up in the first week (0 = almost none, 1 = most of it). Use it to read the 1-week estimates: a low number means prices drift toward the target slowly, so don't over-read a soft week-1.
50 Published records How many dated end-of-day verdicts exist for this book — the length of the public track record you can browse in the Archive below.